【作者】 孟红霞；

【导师】 白学军；

【作者基本信息】 天津师范大学 ， 发展与教育心理学， 2012， 博士

【摘要】 在阅读过程中,眼睛何时(when)移动和移向何处(where)是眼动控制所涉及到的两个基本问题。其中眼睛移向何处(即眼跳行为)是由低水平的视觉线索决定,还是由高水平语言因素决定?目前还处于争论状态,表现为眼动控制模型的不同。在已有研究的基础上,本文试图探讨中文阅读过程中眼跳行为的特点。本论文包含两项研究,共5个实验。研究一主要探讨中文阅读过程中是否存在偏向注视位置,读者眼跳选择的目标,以及低水平的视觉线索因素是否影响读者的眼跳行为。研究二在研究一的基础上,进一步探索高水平语言因素对阅读过程中眼跳行为的影响。具体内容和主要研究结论如下：研究一包含3个实验。实验1,以含有目标词为双字词的句子为实验材料,考察偏向注视位置是否存在,读者的眼跳选择目标,以及笔画数对中文阅读注视位置的影响。实验采用2(首字笔画：多、少)×2(尾字笔画：多、少)的被试内设计。让读者阅读包含四种类型目标词的实验句,并记录被试的眼动轨迹。实验2,考察汉字结构对注视位置的影响。实验采用单因素(汉字结构：左右结构、上下结构)被试内设计。实验3,在实验1和2的基础上,进一步以词切分模糊词组(如“集体力量”)为目标区域,采用阴影标记形成四种呈现方式：正常无阴影条件、词间阴影条件、模糊条件和字间阴影条件,考察词边界信息对词切分模糊词组注视位置效应是否有影响。在研究一的基础上,研究二进一步探讨高水平语言因素是否对中文阅读的注视位置效应产生影响。分为两个实验。实验4,采用单因素(合成词结构类型：并列、偏正)被试内设计,严格控制无关变量,考察合成词的结构类型对注视位置效应的影响。实验5在实验4的基础上,采用单因素(合成词结构类型：并列、偏正)被试内设计,进一步考察同首词素异结构(如“尘垢”和“尘封”)对注视位置效应的影响,进一步验证实验4的结果。综合两项研究的结果,可以得出以下结论：(1)在不同的注视事件中,中文读者的眼动行为表现出分离的现象。具体表现为,当读者对目标词只有一次注视的情况下,首次注视更倾向于落在词的中间位置；当读者对目标词有两次及两次以上注视的情况下,首次注视更多地落在词首位置。表明在中文阅读过程中,单次注视条件下存在偏向注视位置。(2)在多次注视条件下,当读者的首次注视落在词首位置时,再注视该词的概率最高,并且再注视更倾向于落在词尾的位置,表明中文读者的再注视模式比较清晰,有规律。(3)中文读者阅读过程中,对目标词所有向前的注视位置分布并非是一条平行于x轴的平滑曲线,而是在词尾有明显的下降趋势,结果支持眼睛移向何处的决定是以“词”为基础,词是读者下一次眼跳选择的目标,进一步验证了词在中文阅读过程中的重要作用。(4)与拼音文字的结果相似,词的低水平视觉线索——笔画数是中文阅读过程中眼睛移向何处的主要影响因素。具体表现为,当首字为多笔画字时,首次注视倾向于落在首字位置；当首字为少笔画字时,首次注视更多地落在双字词的尾字上。从读者对副中央凹词汇的预加工深度的角度对结果进行了解释。(5)汉字结构、词边界信息等线索不影响中文读者的眼跳行为。此外,词汇的高水平语言因素——合成词结构亦不影响中文阅读过程中眼睛移向何处的决定。更多还原

【Abstract】 Two topics are discussed in the study of eye movement control:what determines when to move the eyes, and what determines where to move during reading? Is the decision of where to move controlled by low-level visual properties or guided by high-level language processing? It is still under debate. Based on the previous studies, we tried to investigate the properties of eye movement behavior during Chinese reading.Two studies including five experiments were carried out. In Study1, we examined whether there was a preferred viewing location, the targeting of readers’saccades, and whether low-level visual properties influenced readers’eye movements behavior. In Study2, on the basis of Study1, whether the high-level lexical properties influenced the landing positions was further explored. The detailed contents and main findings were shown as follows.In Study1, three experiments were included. In Experiment1, sentences including two-character target words were adopted to examine whether the number of stroke influenced the landing positions. The number of stroke of the first and the second constituent characters were manipulated. Participants’eye movements were monitored as they read texts. In Experiment2, on the basis of Experiment1, we explored whether the structure of Chinese characters influenced readers’eye movements during Chinese reading. The structure of Chinese characters (left-right and top-bottom) was manipulated. In Experiment3, the spatially ambiguous words were used as target words to investigate whether the word boundary information influenced the landing positions. The spatially ambiguous words are formed by two consistent Chinese words in which the second constituent character of the first word and the first constituent character of the second word could also form a real Chinese word but it does not semantically related to any of the two words. Highlighting was used to create four analogous conditions:normal Chinese text, text with highlighting used to mark words, text with highlighting to mark the word formed by the second constituent character of the first word and the first constituent character of the second word, and text with highlighting to mark each character.In Study2, two experiments were included to explore whether the high-level lexical properties (the semantic relation between the two constituent characters of a compound word) influenced the landing positions. In Experiment4, we manipulated the constructions of two-character compound words (coordinate and attributive structure). On the basis of Experiment4, in Experiment5, compound words pairs shared the first constituent character but of different constructions (coordinate or attributive) were used as target words to further investigate the influence on landing positions of compound words constructions.To summarize, following conclusions were made.(1) There was different eye movements behavior in different fixation cases. When there was only one fixation on target word, the first fixation mostly landed on the word’ centre. While there were multiple fixations, readers first fixated at the beginning of the target words. There was a preferred viewing location in single-fixation cases during Chinese reading.(2) In multiple fixation cases, if the first fixation landed at the beginning of a target word, the probability of refixating this word was the highest, and refixations were tend to land at the end of the word. The results showed that the refixation pattern of Chinese readers was regular.(3) All forward saccade landing positions distribution curve was not parallel to the x coordinate axes. Especially, the probability of landing the eyes at the end of words decreased. This suggested that the decision about where to move the eyes during Chinese reading is word based, that is, a word object is selected as the next saccade target of Chinese readers. Words play an important role in Chinese text reading.(4) Similar to alphabetic languages, the low-level visual properties of Chinese words (the number of stroke of Chinese characters) influenced the landing positions during Chinese reading. When the number of stroke of the first constituent character was high, the first fixations tend to land on the first character, on the contrary, when the number is low, Chinese readers tended to fixate on the second character.(5) Low-level visual properties of Chinese words, such as character structures and word boundary, did not influence readers’saccades landing positions during reading. Similarly, the decision about where to move the eyes was not influenced by the high-level lexical property, characters’semantic relations of compound words in this case, as well.更多还原